我遇到这个问题已经有很长一段时间了，我找不到解决方案，我尝试了很多事情，但我迷失了方向，但经过一些研究，它似乎与我的处理器有关(可能太旧了？(，我今年才开始为一个项目进行机器学习，所以我根本没有的经验
系统规格
-AMD Phenom II X4 955
-华硕M5A78L-M/USB3(AM3R2(
-EVGA NVIDIA GeForce GT 740(2Gb(
-14 Gb RAM
-Windows 10 Pro
-CUDA套件10.1
-NVIDIA驱动程序441.20-Tensorflow 2.0.0=Tensorflow _gpu-2.0.0-cp37-cp37m-win_amd64(为不带AVX 的遗留cpu构建
我在keras的网页上运行一个卷积滤波器可视化的示例脚本：
from __future__ import print_function
import time
import numpy as np
from PIL import Image as pil_image
from keras.preprocessing.image import save_img
from keras import layers
from keras.applications import vgg16
from keras import backend as K
def normalize(x):
"""utility function to normalize a tensor.
# Arguments
x: An input tensor.
# Returns
The normalized input tensor.
"""
return x / (K.sqrt(K.mean(K.square(x))) + K.epsilon())
def deprocess_image(x):
"""utility function to convert a float array into a valid uint8 image.
# Arguments
x: A numpy-array representing the generated image.
# Returns
A processed numpy-array, which could be used in e.g. imshow.
"""
# normalize tensor: center on 0., ensure std is 0.25
x -= x.mean()
x /= (x.std() + K.epsilon())
x *= 0.25
# clip to [0, 1]
x += 0.5
x = np.clip(x, 0, 1)
# convert to RGB array
x *= 255
if K.image_data_format() == 'channels_first':
x = x.transpose((1, 2, 0))
x = np.clip(x, 0, 255).astype('uint8')
return x
def process_image(x, former):
"""utility function to convert a valid uint8 image back into a float array.
Reverses `deprocess_image`.
# Arguments
x: A numpy-array, which could be used in e.g. imshow.
former: The former numpy-array.
Need to determine the former mean and variance.
# Returns
A processed numpy-array representing the generated image.
"""
if K.image_data_format() == 'channels_first':
x = x.transpose((2, 0, 1))
return (x / 255 - 0.5) * 4 * former.std() + former.mean()
def visualize_layer(model,
layer_name,
step=1.,
epochs=15,
upscaling_steps=9,
upscaling_factor=1.2,
output_dim=(412, 412),
filter_range=(0, None)):
"""Visualizes the most relevant filters of one conv-layer in a certain model.
# Arguments
model: The model containing layer_name.
layer_name: The name of the layer to be visualized.
Has to be a part of model.
step: step size for gradient ascent.
epochs: Number of iterations for gradient ascent.
upscaling_steps: Number of upscaling steps.
Starting image is in this case (80, 80).
upscaling_factor: Factor to which to slowly upgrade
the image towards output_dim.
output_dim: [img_width, img_height] The output image dimensions.
filter_range: Tupel[lower, upper]
Determines the to be computed filter numbers.
If the second value is `None`,
the last filter will be inferred as the upper boundary.
"""
def _generate_filter_image(input_img,
layer_output,
filter_index):
"""Generates image for one particular filter.
# Arguments
input_img: The input-image Tensor.
layer_output: The output-image Tensor.
filter_index: The to be processed filter number.
Assumed to be valid.
#Returns
Either None if no image could be generated.
or a tuple of the image (array) itself and the last loss.
"""
s_time = time.time()
# we build a loss function that maximizes the activation
# of the nth filter of the layer considered
if K.image_data_format() == 'channels_first':
loss = K.mean(layer_output[:, filter_index, :, :])
else:
loss = K.mean(layer_output[:, :, :, filter_index])
# we compute the gradient of the input picture wrt this loss
grads = K.gradients(loss, input_img)[0]
# normalization trick: we normalize the gradient
grads = normalize(grads)
# this function returns the loss and grads given the input picture
iterate = K.function([input_img], [loss, grads])
# we start from a gray image with some random noise
intermediate_dim = tuple(
int(x / (upscaling_factor ** upscaling_steps)) for x in output_dim)
if K.image_data_format() == 'channels_first':
input_img_data = np.random.random(
(1, 3, intermediate_dim[0], intermediate_dim[1]))
else:
input_img_data = np.random.random(
(1, intermediate_dim[0], intermediate_dim[1], 3))
input_img_data = (input_img_data - 0.5) * 20 + 128
# Slowly upscaling towards the original size prevents
# a dominating high-frequency of the to visualized structure
# as it would occur if we directly compute the 412d-image.
# Behaves as a better starting point for each following dimension
# and therefore avoids poor local minima
for up in reversed(range(upscaling_steps)):
# we run gradient ascent for e.g. 20 steps
for _ in range(epochs):
loss_value, grads_value = iterate([input_img_data])
input_img_data += grads_value * step
# some filters get stuck to 0, we can skip them
if loss_value <= K.epsilon():
return None
# Calculate upscaled dimension
intermediate_dim = tuple(
int(x / (upscaling_factor ** up)) for x in output_dim)
# Upscale
img = deprocess_image(input_img_data[0])
img = np.array(pil_image.fromarray(img).resize(intermediate_dim,
pil_image.BICUBIC))
input_img_data = np.expand_dims(
process_image(img, input_img_data[0]), 0)
# decode the resulting input image
img = deprocess_image(input_img_data[0])
e_time = time.time()
print('Costs of filter {:3}: {:5.0f} ( {:4.2f}s )'.format(filter_index,
loss_value,
e_time - s_time))
return img, loss_value
def _draw_filters(filters, n=None):
"""Draw the best filters in a nxn grid.
# Arguments
filters: A List of generated images and their corresponding losses
for each processed filter.
n: dimension of the grid.
If none, the largest possible square will be used
"""
if n is None:
n = int(np.floor(np.sqrt(len(filters))))
# the filters that have the highest loss are assumed to be better-looking.
# we will only keep the top n*n filters.
filters.sort(key=lambda x: x[1], reverse=True)
filters = filters[:n * n]
# build a black picture with enough space for
# e.g. our 8 x 8 filters of size 412 x 412, with a 5px margin in between
MARGIN = 5
width = n * output_dim[0] + (n - 1) * MARGIN
height = n * output_dim[1] + (n - 1) * MARGIN
stitched_filters = np.zeros((width, height, 3), dtype='uint8')
# fill the picture with our saved filters
for i in range(n):
for j in range(n):
img, _ = filters[i * n + j]
width_margin = (output_dim[0] + MARGIN) * i
height_margin = (output_dim[1] + MARGIN) * j
stitched_filters[
width_margin: width_margin + output_dim[0],
height_margin: height_margin + output_dim[1], :] = img
# save the result to disk
save_img('vgg_{0:}_{1:}x{1:}.png'.format(layer_name, n), stitched_filters)
# this is the placeholder for the input images
assert len(model.inputs) == 1
input_img = model.inputs[0]
# get the symbolic outputs of each "key" layer (we gave them unique names).
layer_dict = dict([(layer.name, layer) for layer in model.layers[1:]])
output_layer = layer_dict[layer_name]
assert isinstance(output_layer, layers.Conv2D)
# Compute to be processed filter range
filter_lower = filter_range[0]
filter_upper = (filter_range[1]
if filter_range[1] is not None
else len(output_layer.get_weights()[1]))
assert(filter_lower >= 0
and filter_upper <= len(output_layer.get_weights()[1])
and filter_upper > filter_lower)
print('Compute filters {:} to {:}'.format(filter_lower, filter_upper))
# iterate through each filter and generate its corresponding image
processed_filters = []
for f in range(filter_lower, filter_upper):
img_loss = _generate_filter_image(input_img, output_layer.output, f)
if img_loss is not None:
processed_filters.append(img_loss)
print('{} filter processed.'.format(len(processed_filters)))
# Finally draw and store the best filters to disk
_draw_filters(processed_filters)
if __name__ == '__main__':
# the name of the layer we want to visualize
# (see model definition at keras/applications/vgg16.py)
LAYER_NAME = 'block5_conv1'
# build the VGG16 network with ImageNet weights
vgg = vgg16.VGG16(weights='imagenet', include_top=False)
print('Model loaded.')
vgg.summary()
# example function call
visualize_layer(vgg, LAYER_NAME)
运行代码后，我得到了以下回溯：
Using TensorFlow backend.
2019-11-14 20:13:01.475103: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'cudart64_101.dll'; dlerror: cudart64_101.dll not found
2019-11-14 20:13:05.119364: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library nvcuda.dll
2019-11-14 20:13:05.296509: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1618] Found device 0 with properties:
name: GeForce GT 740 major: 3 minor: 0 memoryClockRate(GHz): 1.202
pciBusID: 0000:01:00.0
2019-11-14 20:13:05.296858: I tensorflow/stream_executor/platform/default/dlopen_checker_stub.cc:25] GPU libraries are statically linked, skip dlopen check.
2019-11-14 20:13:05.304775: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1746] Adding visible gpu devices: 0
2019-11-14 20:13:05.309845: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1618] Found device 0 with properties:
name: GeForce GT 740 major: 3 minor: 0 memoryClockRate(GHz): 1.202
pciBusID: 0000:01:00.0
2019-11-14 20:13:05.310274: I tensorflow/stream_executor/platform/default/dlopen_checker_stub.cc:25] GPU libraries are statically linked, skip dlopen check.
2019-11-14 20:13:05.318573: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1746] Adding visible gpu devices: 0
Traceback (most recent call last):
File "g:Mi unidadCode testinNewTest.py", line 247, in <module>
vgg = vgg16.VGG16(weights='imagenet', include_top=False)
File "C:UsersSEBAAppDataLocalPackagesPythonSoftwareFoundation.Python.3.7_qbz5n2kfra8p0LocalCachelocal-packagesPython37site-packageskerasapplications__init__.py", line 20, in wrapper
return base_fun(*args, **kwargs)
File "C:UsersSEBAAppDataLocalPackagesPythonSoftwareFoundation.Python.3.7_qbz5n2kfra8p0LocalCachelocal-packagesPython37site-packageskerasapplicationsvgg16.py", line 11, in VGG16
return vgg16.VGG16(*args, **kwargs)
File "C:UsersSEBAAppDataLocalPackagesPythonSoftwareFoundation.Python.3.7_qbz5n2kfra8p0LocalCachelocal-packagesPython37site-packageskeras_applicationsvgg16.py", line 112, in VGG16
name='block1_conv1')(img_input)
File "C:UsersSEBAAppDataLocalPackagesPythonSoftwareFoundation.Python.3.7_qbz5n2kfra8p0LocalCachelocal-packagesPython37site-packageskerasbackendtensorflow_backend.py", line 75, in symbolic_fn_wrapper
return func(*args, **kwargs)
File "C:UsersSEBAAppDataLocalPackagesPythonSoftwareFoundation.Python.3.7_qbz5n2kfra8p0LocalCachelocal-packagesPython37site-packageskerasenginebase_layer.py", line 463, in __call__
self.build(unpack_singleton(input_shapes))
File "C:UsersSEBAAppDataLocalPackagesPythonSoftwareFoundation.Python.3.7_qbz5n2kfra8p0LocalCachelocal-packagesPython37site-packageskeraslayersconvolutional.py", line 141, in build
constraint=self.kernel_constraint)
File "C:UsersSEBAAppDataLocalPackagesPythonSoftwareFoundation.Python.3.7_qbz5n2kfra8p0LocalCachelocal-packagesPython37site-packageskerasenginebase_layer.py", line 279, in add_weight
weight = K.variable(initializer(shape, dtype=dtype),
File "C:UsersSEBAAppDataLocalPackagesPythonSoftwareFoundation.Python.3.7_qbz5n2kfra8p0LocalCachelocal-packagesPython37site-packageskerasinitializers.py", line 227, in __call__
dtype=dtype, seed=self.seed)
File "C:UsersSEBAAppDataLocalPackagesPythonSoftwareFoundation.Python.3.7_qbz5n2kfra8p0LocalCachelocal-packagesPython37site-packageskerasbackendtensorflow_backend.py", line 4357, in random_uniform
shape, minval=minval, maxval=maxval, dtype=dtype, seed=seed)
File "C:UsersSEBAAppDataLocalPackagesPythonSoftwareFoundation.Python.3.7_qbz5n2kfra8p0LocalCachelocal-packagesPython37site-packagestensorflow_corepythonkerasbackend.py", line 5598, in random_uniform
shape, minval=minval, maxval=maxval, dtype=dtype, seed=seed)
File "C:UsersSEBAAppDataLocalPackagesPythonSoftwareFoundation.Python.3.7_qbz5n2kfra8p0LocalCachelocal-packagesPython37site-packagestensorflow_corepythonopsrandom_ops.py", line 237, in random_uniform
shape = tensor_util.shape_tensor(shape)
File "C:UsersSEBAAppDataLocalPackagesPythonSoftwareFoundation.Python.3.7_qbz5n2kfra8p0LocalCachelocal-packagesPython37site-packagestensorflow_corepythonframeworktensor_util.py", line 964, in shape_tensor
return ops.convert_to_tensor(shape, dtype=dtype, name="shape")
File "C:UsersSEBAAppDataLocalPackagesPythonSoftwareFoundation.Python.3.7_qbz5n2kfra8p0LocalCachelocal-packagesPython37site-packagestensorflow_corepythonframeworkops.py", line 1184, in convert_to_tensor
return convert_to_tensor_v2(value, dtype, preferred_dtype, name)
File "C:UsersSEBAAppDataLocalPackagesPythonSoftwareFoundation.Python.3.7_qbz5n2kfra8p0LocalCachelocal-packagesPython37site-packagestensorflow_corepythonframeworkops.py", line 1242, in convert_to_tensor_v2
as_ref=False)
File "C:UsersSEBAAppDataLocalPackagesPythonSoftwareFoundation.Python.3.7_qbz5n2kfra8p0LocalCachelocal-packagesPython37site-packagestensorflow_corepythonframeworkops.py", line 1296, in internal_convert_to_tensor
ret = conversion_func(value, dtype=dtype, name=name, as_ref=as_ref)
File "C:UsersSEBAAppDataLocalPackagesPythonSoftwareFoundation.Python.3.7_qbz5n2kfra8p0LocalCachelocal-packagesPython37site-packagestensorflow_corepythonframeworkconstant_op.py", line 286, in _constant_tensor_conversion_function
return constant(v, dtype=dtype, name=name)
File "C:UsersSEBAAppDataLocalPackagesPythonSoftwareFoundation.Python.3.7_qbz5n2kfra8p0LocalCachelocal-packagesPython37site-packagestensorflow_corepythonframeworkconstant_op.py", line 227, in constant
allow_broadcast=True)
File "C:UsersSEBAAppDataLocalPackagesPythonSoftwareFoundation.Python.3.7_qbz5n2kfra8p0LocalCachelocal-packagesPython37site-packagestensorflow_corepythonframeworkconstant_op.py", line 235, in _constant_impl
t = convert_to_eager_tensor(value, ctx, dtype)
File "C:UsersSEBAAppDataLocalPackagesPythonSoftwareFoundation.Python.3.7_qbz5n2kfra8p0LocalCachelocal-packagesPython37site-packagestensorflow_corepythonframeworkconstant_op.py", line 95, in convert_to_eager_tensor
ctx.ensure_initialized()
File "C:UsersSEBAAppDataLocalPackagesPythonSoftwareFoundation.Python.3.7_qbz5n2kfra8p0LocalCachelocal-packagesPython37site-packagestensorflow_corepythoneagercontext.py", line 492, in ensure_initialized
self._context_handle = pywrap_tensorflow.TFE_NewContext(opts)
tensorflow.python.framework.errors_impl.InternalError: cudaGetDevice() failed. Status: cudaGetErrorString symbol not found.
我还运行：
import tensorflow as tf
print("Num GPUs Available: ",
len(tf.config.experimental.list_physical_devices('GPU')))
获得：
2019-11-14 20:46:58.806642: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'cudart64_101.dll'; dlerror: cudart64_101.dll not found
2019-11-14 20:47:02.876517: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library nvcuda.dll
2019-11-14 20:47:02.914412: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1618] Found device 0 with properties:
name: GeForce GT 740 major: 3 minor: 0 memoryClockRate(GHz): 1.202
pciBusID: 0000:01:00.0
2019-11-14 20:47:02.914794: I tensorflow/stream_executor/platform/default/dlopen_checker_stub.cc:25] GPU libraries are statically linked, skip dlopen check.
2019-11-14 20:47:02.915417: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1746] Adding visible gpu devices: Num GPUs Available: 1
0
我希望有人能帮忙，我真的需要这个感谢
CUDA似乎不支持您的GPU，因此TensorFlow出现问题(因此TensorFlow无法识别您的GPU(。
根据NVIDIA的GPU文档网站，您的GPU的计算兼容性为3.0。然而，根据TensorFlow的文档，需要一个计算兼容性>=3.5的CUDA设备。
我建议使用TensorFlow与CPU而不是GPU。
谢谢！
根据NVIDIA的GPU文档网站，您的GPU计算兼容性为3.0。然而，根据TensorFlow的文档，需要一个计算兼容性>=3.5的CUDA设备。
阅读后，我开始研究并发现：https://medium.com/@mccann.matt/compiling-传感器流量-带cuda-3-0-支架-42d8fe0bf3b5关于如何编译与计算能力3.0兼容的tensorflow版本，这可能很痛苦，但我想使用GPU来训练具有非常大数据集的CNN，所以我认为值得一试。如果它有效与否，我会发布一个更新
感谢您指出：(